Here's a brief sketch of a philosophical argument for a fairly controversial thesis: that any political group with power that operates by consensus will opt to increase their own power rather than work towards ethical goals, even if that group is composed entirely of ethical individuals. I don't expect this sketch to be a solid proof, nor do I expect it necessarily to be convincing, but I think there's a solid argument somewhere in here for an individualist ethics.
1) Let's take a pretty simple psychological model for this thought experiment. Imagine that the people in this group have one main goal: to use the power of the group to achieve a particular ethical outcome. Let's call the first person in the group P1, and the ethical outcome that they want E1. Likewise, P2 wants E2, and so on.
2) Now for P1 to achieve E1 most optimally - let's say that E1 is 'increase social justice' - he needs two things: firstly, the group to agree with him, and secondly, the group to have sufficient power to carry out policies that would lead to E1.
3) Since P2 wishes to achieve E2 most optimally - let's say that E2 is 'provide resources to the poor' - he needs two things as well. He needs the group to assent to E2, and he needs the group to have sufficient power to carry out policies that would lead to E2.
4) So when P1 and P2 get together, they will come into conflict over their particular ethical goals, but be in perfect harmony concerning the subsidiary goal of increasing the group's power. Therefore, they will increase the power of the group far more easily and efficiently than they will use that power to go after any particular ethical goal.
5) When you add more people to the group, the conflict over the particular ethical goal worsens, but the goal of increasing the group's power will always be shared by all. Therefore, the larger the group, the less likely it is that an effective compromise will be reached. Groups of very large size would tend to be paralyzed giants: groups with huge political power that do not actually use that power.
So, assuming my simplified psychological model, I've shown that political groups tend to grow in power much more quickly than they move towards ethical goals.
What are the problems with my argument?
1) Let's start with the obvious: I'm assuming that rational, ethical actors won't be able to compromise. P1 and P2 might well get together and find policies that work towards both their goals simultaneously.
2) I'm also assuming that the people in the group want to increase their power constantly, rather than wanting to increase it only once a consensus has been reached.
3) Groups can be formed on the basis of a particular ethical goal - to reduce suffering in the Sudan, for instance - and in that case there's no reason to believe the members of the group would be in conflict.
Why do I think my argument's worthwhile, despite those problems?
Well, I think that this is how politics works in the real world: groups with power tend to accrue other powers more reliably than they tend to accomplish any other goal. This is in line with my anarchism, obviously, and other people are free to disagree. If I'm right, then there's an argument that can be made along these lines successfully, even if it's not the argument I outlined above.
My thoughts on this matter are still pretty much in their infancy, so I would appreciate any criticism from other sources.